News for hackers. Stuff that matters.
Re:Show HN: High-performance ahead-of-time compiler for Mach... by marmaduke on Thursday December 13, 2018 @12:01PM (#18673852)
how tied are the compiler aspects to the specific kernels (eg for convolutional layers)? Could load balancing and fusion logic be broken out into a library which could work for user defined kernels?
Re:Show HN: High-performance ahead-of-time compiler for Mach... by andrew-wja on Thursday December 13, 2018 @01:15PM (#18674603)
They aren't tied at all -- in fact the optimizer is a totally separate project (triNNity-optimizer) that just does graph optimization. You can add user defined kernels as long as you have some way of microbenchmarking them!